|
(11) | EP 2 408 096 A1 |
| (12) | EUROPEAN PATENT APPLICATION |
|
|
|
|
||||||||||||||||||||
| (54) | Current-fed converter with quadratic conversion ratio |
| (57) A converter circuit comprising first and second input terminals (1, 2) for receiving
input current from a current source, first capacitor (C1) connected between the first and second input terminals, second capacitor (C2), the first terminal of which is connected to the second input terminal (2) and the
second terminal of which forms a positive voltage node (3), first and third semiconductor
components (S1, S3) connected in series between the first input terminal (1) and positive voltage node
(3), the midpoint between the series connection forming a first node (4), series connection
of first inductive component (L1), first diode (D1) and second inductive component (L2), which series connection is connected between the second input terminal (2) and
first node (4), the polarity of the first diode being such that it allows a current
to pass from the direction of the second input terminal; the point between the first
inductive component (L1) and the first diode (D1) forming a second node (5) and the point between the first diode (D1) and the second inductive component forming a third node (6), second diode (D2) connected between the second node (5) and the first node (4) in a manner which allows
current to pass from the direction of the second node (5) to the first node (4), third
capacitor (C3) connected between first input terminal (1) and the third node (6), second and fourth
semiconductor components (S2, S4) connected in series in parallel with the series connection first and third semiconductor
components (S1, S3), third inductive component (L3), the first end of which is connected to a fourth node (7) formed between the second
and fourth semiconductor components (S2, S4) and the second end of which produces first output terminal (8), the second output
terminal (9) being formed of the first input terminal (1), the first and the third
semiconductor components (S1, S3) being arranged to control the voltage (Uin) between the first and second input terminals (1, 2).
|
FIELD OF THE INVENTION
[0001] The present invention relates to converters, and more particularly to current-fed converters having semiquadratic properties.
BACKGROUND OF THE INVENTION
[0002] Converters are common electric apparatuses that are used for converting electricity from one form to another. Converters are typically used for rising or lowering voltage or current levels for adapting the voltage of a source to the voltage of the load.
[0003] Converters are also used in connection photovoltaic (PV) panels for supplying the power available from panel. The power may be supplied directly to an alternating voltage network, DC -voltage bus or to a power consuming load. Depending on the amount of PV cells or panels connected for supplying power, the voltage is either rised or lowered for supplying the load with appropriate voltage.
[0004] In connection with the PV systems the converter system also keeps the operating point of the panel at its maximum power point. A PV panel has a certain operating point at which the extracted power has its maximum. A maximum power point tracker (MPPT) calculates the operating point from measured current and voltage of the panel and gives a reference value for either output voltage or current of the panel. The control system of the converter controls the actual output value of the panel to the reference and the maximum available power is obtained from the panel. The MPPT calculates continuously the reference value and the control system adapts to this situation enabling continuous operation in the maximum power point regardless of the varying operating conditions.
[0005] Converter topologies can be roughly divided into current fed and voltage fed converters. In voltage fed converters the power source feeds a voltage to the input of the converter and in current fed converters the source feeds current to the input. Due to the constant-current nature of PV panels, the optimal interfacing of the panels can be implemented by using current fed converters. A current fed converter can operate within the whole range of the Ul curve of the panel from the short-circuit to open-circuit conditions. In voltage-fed converter the operation can only be carried out at the voltages equal or higher than the maximum power point voltage.
BRIEF DESCRIPTION OF THE INVENTION
[0006] An object of the present invention is to provide a current fed converter structure having low losses and wide input voltage range. The object of the invention is achieved by a converter which is characterized by what is stated in the independent claim. The preferred embodiments of the invention are disclosed in the dependent claims.
[0007] The invention is based on the idea of using a current fed converter structure having a novel topology. The converter of the invention has a wide conversion ratio producing semiquadratic current characteristics in view of the duty ratio of the converter. The topology of the converter can be used to extract the maximum power from a power source, such a photovoltaic panel. The topology can be used for supplying voltage to a DC bus or with back end inverter to an alternating voltage grid.
[0008] One of the advantages of the topology of the converter is that when connected via inverter part to the grid, the power fluctuations at twice the grid frequency are not visible at the input terminals of the converter, thus reducing the problems relating to the fluctuation of the PV panel operating point.
BRIEF DESCRIPTION OF THE DRAWINGS
[0009] In the following the invention will be described in greater detail by means of preferred embodiments with reference to the attached [accompanying] drawings, in which
Figure 1 illustrates the main circuit of the converter of the invention;
Figure 2 shows the converter of Figure 1 with current and voltage definitions;
Figure 3 shows the on-time circuit structure of the converter;
Figure 4 shows the off-time circuit structure of the converter;
Figure 5 shows approximate waveforms of voltages and currents;
Figure 6 shows the on-off-time circuit structure of the converter;
Figure 7 shows the off-on-time circuit structure of the converter;
Figure 8 shows an example of duty ratio and switch control signal generation;
Figure 9 shows approximate waveforms when d1 > d2;
Figure 10 shows approximate waveforms when d1 < d2;
Figure 11 shows main circuit setup of the converter according to an embodiment;
Figure 12 shows switching frequency averaged capacitor voltages at the grid connected operation;
Figure 13 shows switching frequency averaged inductor currents at the grid connected operation; and
Figure 14 shows waveforms of the grid variables.
DETAILED DESCRIPTION OF THE INVENTION
[0010] Figure 1 shows the circuit diagram of the current-fed semiquadratic buck-boost converter of the invention. In Figure 1 a current source is connected between first and second input terminals 1, 2 in parallel with capacitor C1. Another capacitor C2 is connected to the second input terminal and the other terminal of the capacitor C2 forms a positive voltage node 3.
[0011] The converter further comprises a series connection of first and third semiconductor components S1, S3. The series connection is connected between the first input terminal 1 and the positive voltage node 3. The first semiconductor component is either a controlled switch component, such as FET-component, or a diode. The diode D3 is illustrated in Figure 1 in dashed line and as mentioned, diode D3 can substitute the switch component S1. The third semiconductor component is a controlled switch component. The polarities of the components are such that when turned on, the current can flow through the components from the positive voltage node towards the first input terminal, thus the voltage from the positive voltage node is blocked.
[0012] The midpoint between the semiconductor components S1 and S3 is denoted as the first node 4 for the simplicity of the description.
[0013] A series connection of first inductive component L1, first diode D1 and second inductive component L2 is connected between the second input terminal 2 and the first node 4. The diode allows the current to pass from the direction of the second input terminal 2. A second diode D2 is connected in parallel with the series connection of the first diode D1 and the second inductor L2 (i.e. from node 5 to node 4). The polarity of the second diode is similar to that of the first diode, and current can pass from the second node 5 to the first node 4.
[0014] The converter comprises further a third capacitor C3, which is connected between the first input terminal 1 and the third node 6, which is the connection point between the first diode D1 and the second inductor L2. Further, the converter of Figure 1 comprises second and fourth semiconductor components S2, S4 connected in series. The series connection is connected in parallel with the first and third semiconductor components with similar polarities. As with the first and third semiconductors, the both components S2 and S4 may be controlled switches. As indicated in Figure 1, the fourth switch may be replaced with a diode D4.
[0015] A third inductor L3 is connected to the point between the second and fourth semiconductor components and this point is also referred to as fourth node 7. The other end of the third inductor forms a first output terminal and the second output terminal has the same potential as the first input terminal 1.
[0016] According to the invention, the first and third semiconductor components are arranged to control the voltage uin between the first and second input terminals 1 and 2 i.e. the input voltage. Since the converter is a current fed converter, the input current to the converter depends on the supplying source and the voltage is controlled with the switches to the voltage that suits the operation best.
[0017] The current-fed semiquadratic buck-boost converter shown in Fig. 1 has steady-state conversion ratios such that the input current (Iin) is reflected to the output (lo) multiplied by the square of the duty ratio and divided by the complement of the duty ratio (i.e., M(D) = D2 / D' ) and the output voltage (Uo) to the input (Uin) with the same conversion ratio. The converter is a combination of buck (S1, S3) and boost (S2, S4) type-converters which can be controlled either separately (dual PWM mode) or together (single PWM mode). In the following, the operation of the converter is first studied in the single PWM mode operation and then in the dual PWM mode operation.
Single PWM mode
[0018] The power stage of the current-fed semiquadratic buck-boost converter of the invention is shown in Figure 2 with the defined components and the polarities of relevant currents and voltages. As mentioned above, the on-time switches S1 and S4 can be substituted also with diodes D3 and D4.
[0019] During the on-time, the switches S2 and S3 and the diode D2 are off and the switches S1 and S4 and the diode D1 are conducting yielding the on-time circuit structure given in Figure 3. During the off-time, the switches S2 and S3 and the diode D2 are conducting and the switches S1 and S4 and the diode D1 are off yielding the off-time circuit structure given in Figure 4.
[0022] According to (1) and (2), the average voltages across the inductors, average currents through the capacitors as well as average input voltage, intermediate voltage Uim (i.e. the voltage of the series connection of the capacitors C1 and C2) and output current become as
[0023] In the above set of equations, the subscripts denote the component the current or voltage of which is denoted by that equation. In order to maintain flux linkage and charge balances, the average voltages across the inductors and average currents through the capacitors have to be zero. According to these principles the operating-point-related steady-state variables become with denoting D'=1―D
[0024] which indicates that the input current is reflected to the output multiplied by the square of the duty ratio and divided by its complement (i.e. M(D) = D2 / D' ), and the output voltage to the input with the same conversion ratio. The output voltage is reflected to the intermediate voltage divided by the complement of the duty ratio. The switching frequency of the converter is assumed to be fs and consequently, the cycle time Ts. =1/fs. The duty cycle D = Ton / Toff and therefore, the length of on-time Ton = DTS. and the length of off-time Toff = D'Ts because Ts = Ton +Toff .
[0025] The approximate waveforms of the capacitor voltages and inductor currents are shown in Figure 5. The steady-state values are assumed to be large compared to the peak-to-peak ripple, i.e. only steady-state values affect the ripple slopes. The approximate waveform of the capacitor voltages is shown in Figure 5 as the curve a) exhibiting triangle shapes. The on-time and off-time slopes of the waveform are denoted by m1Ci and ―m2Ci. They can be given for the capacitor C1 by m1C1 = DIin l C1 and m2C1 = D2Iin /(D'C1) as well as for the capacitor C2 by m1C2 = DIin l C2 and m2C2 = D2Iin l (D'C2) and for the capacitor C3 by m1C3 = D'Iin / C3 and m2C3 = DIin / C3 . According to the defined slopes, the peak-to-peak ripples associated to the capacitors can be determined by ΔuCi―pp = m1CiDTs. The average voltages are defined in (4). The selection of the capacitors can be based on the defined peak-to-peak ripple and average voltages.
[0026] The approximate waveform of the inductor currents is shown in Fig. 5 as the curve b) exhibiting also triangle shapes. The on-time and off-time slopes of the waveform are denoted by ―m1Li and m2Li. They can be given for the inductor L1 by m1L1 = DUo/L1 and m2L1 = D2Uo l (D'L1 ) as well as for the inductor L2 by m1L2 = Uo / L2 and m2L2 = DUo/(D'L2) and for the inductor L3 by m1L3 = Uo l L3 and m2L3 = DUo / (D'L3). The corresponding peak-to-peak ripple currents can be given by ΔiLi―pp = m1LiDTs and the corresponding average currents are defined in (4). The selection of the inductors can be based on the defined peak-to-peak ripple and average currents.
[0027] When the converter operated in the single PWM mode feeds a DC voltage bus, the control may be carried out in the following manner. Supposing that the power supply feeding the converter is a photovoltaic panel, a maximum power point tracker is used for extracting the maximum available power. The input current and voltage to the converter are measured. These measurements are fed to MPPT device which outputs a reference for the input voltage. The input voltage reference and the measured voltage are fed to a voltage controller which controls the voltage to the desired value by controlling the pulse ratio of the switches to a suitable value.
[0028] Preferably the output voltage of the converter is limited to a maximum value. If the converter produces higher voltage than the limit, voltage is lowered by changing the operation point of the converter. This can be carried out by changing the input voltage reference so that the converter no longer operates at the maximum power point. To achieve this the measured output voltage and the value of the maximum output voltage are fed to a controller, and once the measured voltage exceeds the limit, the controller outputs a control term that is subtracted from the value produced by the MPPT. Of cource, other measures can also be used for limiting the output voltage to a suitable level.
Dual PWM mode
[0029] In the dual PWM mode duty ratios of the buck and boost switch pairs (S1,S3 and S2,S4 , respectively) are controlled separately, allowing e.g. dedicated control systems for the output current and input voltage as well as separate control of the intermediate voltage. Besides the on- and off-times of the single PWM converter, the circuit can exhibit two extra states denoted as on-off- and off-on-times. During the on-off-time, the switches S3 and S4 and the diode D2 are off and the switches S1 and S2 and the diode D1 are conducting yielding the on-off-time circuit structure given in Fig. 5. During the off-on-time, the switches S3 and S4 and the diode D2 are conducting and the switches S1 and S2 and the diode D1 are off yielding the off-on-time circuit structure given in Fig. 7.
[0032] In order to solve the average voltages across the inductors, average currents through the capacitors as well as average input voltage, intermediate voltage and output current, we assume synchronous PWM generation. The switching frequencies are assumed equal as well as the beginning of the on-times. The synchronous PWM generation is realised either by digital control system or comparing the control signals to the same PWM-ramp in analog modulator as depicted in Fig. 8, where Ga is the modulator related gain 1/Vm and Vm is the peak-to-peak value of the PWM-ramp slope, scaling d1 and d2 between zero and one.
[0033] According to Fig. 8 an extra state appears between on- and off-times when d1 # d2 . If d1 > d2 , the extra state is the on-off-time and if d1 < d2 , the extra state is the off-on-time. Now the needed average values can be solved by common average integrals where Ts is the cycle time. For d1 < d2 we get
[0036] In order to maintain flux linkage and charge balances, the average voltages across the inductors and average currents through the capacitors have to be zero. According to these principles the operating-point-related steady-state variables become with denoting D'2 =1―D2 and D1' =1―D1
[0037] The steady-state operating point solution is similar to the single PWM case, but the duty ratios of the buck- and boost-parts are separated. The switching frequency of the converter is again assumed to be fs and consequently, the cycle time Ts =1 / fs When d1 > d2 the duty cycle D2 = Ton / (Ton-off + Toff) and therefore, the length of on-time Ton = D2Ts and the length of off-time Toff = D1'Ts because Ts =Ton + Ton-off + Toff and Ton-off = D1Ts ― D2Ts. When d1 < d2 the duty cycle D1 = Ton / (Toff-on + Toff) and therefore, the length of on-time Ton = D1Ts and the length of off-time Toff = D'2Ts because Ts = Ton + Toff-on + Toff and Toff-on = D2Ts ― D1Ts .
[0038] The approximate waveforms of the capacitor voltages and inductor currents are shown in Figure 9 and Figure 10 for the corresponding duty ratio relations. The steady-state values are assumed to be large compared to the peak-to-peak ripple, i.e. only steady-state values affect the ripple slopes. The capacitor voltage slopes shown in Fig. 9 and Fig. 10 as the curves a) and b) can be given for the capacitor C1 by m1C1 = D1Iin/C1, m2C1 = D12Iin/(D'2C1) and m3C1 = ( D1 ― D12 / D'2) Iin/C1 as well as for the capacitor C2 by m1C2 = D1Iin /C2 , m2C2 = D12Iin / (D'2C2) and m3C2 = (D1 ― D12 / D'2 )Iin / C2. The capacitor C3 slopes can be given by m1C3 = D'1Iin and m2C3 = D1Iin
[0039] Note that the slopes m3Ci are negative if D1 > D'2. When D1 = D'2, the slopes m3Ci are approximately zero. The capacitor C1 and C2 slopes are approximately zero also during the off-on-time. The peak-to-peak ripples associated to the capacitors can be determined according to the defined slopes. When D1<D'2, ΔuCi―pp = m2CiToff, where Toff = D'1Ts when d1 > d2 and Toff = D'2Ts. when d1 < d2 . When D1 > D'2, ΔuCi―pp = m1CiTon, where Ton = D2Ts when d1 > d2 and Ton = D1Ts when d1 < d2 . The average voltages are defined in (10). The selection of the capacitors can be based on the defined peak-to-peak ripple and average voltages.
[0040] The inductor current slopes shown in Fig. 9 and Fig. 10 as the curves c) and d) can be given for the inductor L1 by m1L1 = D1D'1Uo /(D'2L1) and m2L1 = D12Uo/(D'2L1) as well as for the inductor L2 by m1L2 = D'1Uo / (D'2L2) and m2L2 = D1Uo / (D'2L2). The inductor L3 slopes can be given by m1L3 = Uo / L3 and m2L3 = D2Uo / (D'2L3). The corresponding peak-to-peak ripple currents can be given by ΔiLi = m1LiTon, where Ton = D2Ts when d1 > d2 and Ton = D1Ts when d1 < d2 . The average currents are defined in (10). The selection of the inductors can be based on the defined peak-to-peak ripple and average currents.
[0041] When the converter operated in the dual PWM mode feeds a DC voltage bus, the control may be carried out in the following manner. Supposing that the power supply feeding the converter is a photovoltaic panel, a maximum power point tracker is used for extracting the maximum available power. The input current and voltage to the converter are measured. These measurements are fed to MPPT device which outputs a reference for the input voltage. The input voltage reference and the measured voltage are fed to a voltage controller which controls the voltage to the desired value by controlling the pulse ratio of the switches S1 and S3 to a suitable value.
[0042] As the switches S1, S3 and switches S2, S4 are modulated separately, the output voltage can be kept at a desired level by modulating the switches S2 and S4.
[0043] The above calculations and the waveforms were obtained with synchronized switching periods. It is however clear, that the switches S1, S3 and S2, S4 can be modulated completely independent from each other, meaning that even the modulation frequency may not be the same for both switch pairs.
Grid interfacing
[0044] In the above the converter was described as providing voltage to a DC bus. In the following, the topology of the invention is used for supplying power to alternating grid thereby forming effectively an inverter. The transformerless inverter technology is considered to be a low-cost and feasible solution for interfacing solar generator into the grid. The main problems of such solutions are considered to be the high common-mode currents and the fluctuation of the input power at twice the line frequency causing safety problems and reducing the energy harvesting efficiency.
[0045] The converter according to an embodiment forms a single-phase current-fed solar inverter comprising a current-fed quadratic buck converter S1, S3 and boost converter S2, S4, a line-frequency inverter (S5 ― S8) and an EMI filter as shown in Figure 11.
[0046] The buck converter keeps its input voltage constant at the voltage determined by the MPPT device by applying negative feedback control. The boost converter supplies full-wave rectified sinus shaped output current corresponding to the maximum power the input source (i.e., e.g. solar generator) can provide. The boost converter keeps its input voltage (i.e., the voltage of the sum of the input capacitors) constant at a level needed for supplying the grid current and determined by the input-voltage-monitoring device. The level can be chosen to minimize the losses in the buck and boost converters. The |AC| / AC inverter works in such a way that the switch pair (S5,S7) conducts during the line half cycle when the grid voltage is positive and the switch pair (S6,S8) during the line half cycle when the grid voltage is positive. The switches (S5-S8) can be implemented by several techniques including MOSFET, IGBT, silicon controlled rectifier, etc. The EMI filter, consisting for example by a capacitor and inductor, connected at the output of the |AC |/ AC converter removes the switching-frequency noise from the output current.
[0047] The current-fed semi-quadratic buck-boost converter integrated with a line-frequency inverter is a feasible solution for interfacing solar generator into the single-phase grid without an isolation transformer. The output current of the semi-quadratic buck-boost is shaped to resemble full-wave rectified grid voltage, which polarity is determined by the line-frequency inverter according to the grid voltage. The voltages imposed over parasitic capacitances from the input source to the neutral conductor are ideally free of high frequency noise, efficiently reducing ground leakage currents, which are considered main problems in transformerless grid connected photovoltaic applications. The main circuit setup of the inverter with the basic control functions is shown in Figure 11.
[0048] The common-mode voltage created by the circuitry is free of highfrequency noise and half the grid voltage, which effectively reduces the common mode current. The input-voltage controlled of the buck converter effectively reduces the effect of the power fluctuation at the output of the input source down to zero. The intermediate voltage (uim) has to be controlled to be slightly higher than the peak grid voltage but can be optimized in respect to the power losses in the buck and boost stages. The input voltage of the buck converter has to be less than the intermediate voltage and its minimum level is determined by the minimum usable duty ratio by Uin―min = D2minUDC―max The quadratic nature of the input converter enables the implementation of a module integrated solar inverter.
[0049] The dual PWM mode described above is used due to its advantages over the single PWM mode. The dedicated input voltage control efficiently reduces the effect of the output power fluctuation in the input with a modest increase in complexity. Smaller input capacitance is required while the energy harvesting efficiency is increased. The buck switch pair (S1,S3) is used to keep the input voltage constant at a level determined by the maximum power point tracking (MPPT) device by applying negative feedback control.
[0050] The boost switch pair (S2,S4) is used to deliver the full-wave rectified sinus shaped output current. It has been observed that high control bandwidths are hard to achieve under direct duty ratio control of the output current with current-fed converters. Dynamic analysis (not provided) of the dual PWM semi-quadratic buck-boost converter reveals a right-half-plane (RHP) zero in the control-to-output related transfer functions of the boost-part limiting the control bandwidth. The RHP zero is a characteristic property of a boost converter, whether voltage fed or current fed. A positive feedback output current loop unstabilises the converter but allows high bandwidth current reference tracking while the intermediate voltage uim keeps at adequate levels. Now it can be observed that the converter may be stabilised by another positive feedback loop from the intermediate voltage. After addition of the current loop, the dynamic analysis reveals a RHP pole in the output-current-reference-to-intermediate-voltage transfer function. According to system theory, the converter is stable when the intermediate voltage loop bandwidth exceeds the RHP pole frequency. Naturally, the input voltage controller affects also the dynamics of the boost-part.
[0051] The intermediate voltage has to be kept at a level higher than the peak grid voltage. The grid voltage peak identifier determines the intermediate voltage reference to allow operation with various grid voltages. The level can be chosen to minimise the losses in the converter with a transient margin. The line-frequency inverter works in such a way that the switch pair S5,S8 conducts during the positive half of the grid voltage line cycle and the switch pair S6,S7 conducts during the line cycle half when the grid voltage is negative. Low conduction loss devices are recommended. The EMI filter connected at the output attenuates the switching frequency noise of the output current and ensures the EMC-compatibility of the inverter.
[0052] The control system of the inverter according to an embodiment is described next with reference to Figure 11. A current source 101, such as a photo voltaic panel, string or array, is connected to the input of the inverter. The current source produces input current iin, which is measured together with the input voltage uin. The measured values are fed to a maximum power point tracker, which calculates a reference value for the input voltage uin-ref. When such a voltage is obtained to the input of the inverter, the PV panel is operated at is maximum power point and all available power is extracted from the panel.
[0053] The input voltage reference uin-ref is fed to a PWM controller 103 together with the measured input voltage uin. The controller 103 produces a duty ratio for gate driver 104, which controls the switches S1 and S3 according to the obtained duty ratio for controlling the input voltage to correspond to the reference voltage. The switches S1 and S3 thus effectively control the input voltage to the reference voltage, which is obtained from the measured input current using an MPP algorithm.
[0054] In the inverter of the embodiment the grid voltage ugrid is also measured. The measured voltage value is fed to block 105. Block 105 serves as polarity controller for the inverter which is synchronized to the grid voltage. The polarity controller outputs the information of the polarity of the grid voltage to switch driver 106 which controls switches S5 and S8 conductive when the grid voltage is positive and switches S6 and S7 conductive when the grid voltage is negative.
[0055] Block 105 comprises also a rectifier for obtaining the rectified sinusoidal shape of the grid voltage. The grid voltage is rectified and scaled by multiplying it with a constant k.
[0056] Further, block 105 includes a peak identifier, which identifies the amplitude of the grid voltage, for producing a reference value uim-ref for the intermediate voltage. The intermediate voltage uim is measured and fed to intermediate voltage controller 107 together with the reference value uim-ref. The output of the voltage controller 107 is fed to a multiplier block 108 which multiplies said output and the scaled and rectified value of the grid voltage k|ugrid| for producing a reference for the rectified value of grid current i|AC|-ref. This reference value is fed to grid current controller which also receives measured grid current value i|AC|. The measured value is also rectified value since it is measured before the inverting switches. Grid current controller 109 outputs a pulse ratio for gate driver 110, which controls switches S2 and S4 according to the pulse ratio for keeping the intermediate voltage at its reference value and for producing a current having a rectified sinusoidal shape in synchronism with the grid voltage.
[0057] To easily demonstrate the low frequency behaviour of the inverter in Figure 11, time domain simulations in steady-state have been carried out with a switching frequency averaged model (i.e. the switching frequency ripple is removed from the waveforms). The parameters used in the simulation are listed in Table 1.
Table 1. Simulation parameters.
| um-ref | uim-ref | Ugrid | iin | C1 | C2 | C3 | L1 | L2 |
| 25V | 360V | 230V | 7.5A | 22µF | 440µF | 22µF | 220µH | 1mH |
[0058] Figure 12 shows the capacitor voltages. As can be seen, the input voltage (capacitor C1 voltage, UC1) ripple at twice the grid frequency can be controlled to neglibly small values. The voltage ripple at the intermediate voltage (uC1 + uC2) is similar to the input voltage ripple in any common grid-connected single-phase photovoltaic inverter. The voltage uC3 is scaled down from the intermediate voltage by the duty ratio d1 . The inductor currents are shown in Figure 13. The low-frequency ripple current of the capacitor C1 is steered to the inductor L1 in order to keep the input voltage constant. The ripple current of the inductor L2 is scaled down from the inductor L1 ripple current by the duty ratio d1 . The full-wave-rectified shaped current of the inductor L3 forms the grid current after the line-frequency inverter. The resulting grid current with the grid voltage is shown in Figure 14.
[0059] In order to design the inverter properly, the low frequency ripples needs to be solved. The switching frequency ripples were analysed above in connection with the converter structure. The variables in angle brackets are the switching frequency averaged values. The low frequency ripples can be approximated by calculating the power fluctuation caused by the sinusoidal grid variables. The grid current controller injects sinusoidal current to the grid at unity power factor. The grid angular frequency is denoted by ω0 At steady-state
[0060] According to the trigonometric double-angle formulas sin2 (x) = (1 ― cos(2x)) / 2 . Now it is clearly seen that the power fluctuates at twice the grid frequency. Given as a function of the input power and efficiency of the converter, denoted by η, steady-state grid power is
[0061] Since the input voltage controller keeps the input voltage constant, the average current through the capacitor C1 is zero. Accordingly, the power fluctuation imposed low frequency current appears through the capacitor C2 . At steady-state the dc-component must be zero in order to maintain the charge balance. If the ripple voltage over the capacitor C2 is assumed small compared to the dc-component, we get (13) applying the current directions of Figure 2.
[0062] According to the basic circuit theory, the inductor voltages and capacitor currents can be presented as
and
, respectively.
[0063] The voltage ripple over the capacitor C2' which is also the voltage ripple of the intermediate voltage, is
[0064] Since the input current is constant, 〈iC1〉 = 0 and 〈iL1〉 = 〈iin〉 + 〈iC2〉 - 〈iC1〉 , the ripple current through the inductor L1 is simply
[0065] In order to keep the input voltage constant and power flow towards the grid, the ripple current amplitude of the inductor L1 must not exceed the input current. If the assumption for (13) is justified, we may also assume that the ripple component in the duty ratio d1 is small compared to the dc-component. Now the ripple current of the inductor L2 and the ripple voltage of the capacitor C3 can be given as shown in (16) and (17), respectively.
[0066] It is to be understood, that the converter also comprises means for controlling the semiconductor components. These means comprise suitable drive circuitry able to control the components to a conductive and blocking state. Drive circuitry includes for example auxiliary power sources, modulators and other peripheral circuits. It is also clear that the above mentioned controller for current and voltage control are implemented in a manner known by the skilled person. Further, the used maximum power point tracker may be of any available type as long as it produces a reference value for the input voltage. In the above description referring to drawings some of the features are incorporated in blocks. It is clear that the functionality required for the invention can be structured in functional blocks other than what is described above and in the drawings.
1. A converter circuit comprising
first and second input terminals (1, 2) for receiving input current from a current source,
first capacitor (C1) connected between the first and second input terminals, characterized in that the converter further comprises
second capacitor (C2), the first terminal of which is connected to the second input terminal (2) and the second terminal of which forms a positive voltage node (3),
first and third semiconductor components (S1, S3) connected in series between the first input terminal (1) and positive voltage node (3), the midpoint between the series connection forming a first node (4),
series connection of first inductive component (L1), first diode (D1) and second inductive component (L2), which series connection is connected between the second input terminal (2) and first node (4), the polarity of the first diode being such that it allows a current to pass from the direction of the second input terminal; the point between the first inductive component (L1) and the first diode (D1) forming a second node (5) and the point between the first diode (D1) and the second inductive component forming a third node (6),
second diode (D2) connected between the second node (5) and the first node (4) in a manner which allows current to pass from the direction of the second node (5) to the first node (4),
third capacitor (C3) connected between first input terminal (1) and the third node (6),
second and fourth semiconductor components (S2, S4) connected in series in parallel with the series connection first and third semiconductor components (S1, S3),
third inductive component (L3), the first end of which is connected to a fourth node (7) formed between the second and fourth semiconductor components (S2, S4) and the second end of which produces first output terminal (8), the second output terminal (9) being formed of the first input terminal (1),
the first and the third semiconductor components (S1, S3) being arranged to control the voltage (Uin) between the first and second input terminals (1, 2).
first and second input terminals (1, 2) for receiving input current from a current source,
first capacitor (C1) connected between the first and second input terminals, characterized in that the converter further comprises
second capacitor (C2), the first terminal of which is connected to the second input terminal (2) and the second terminal of which forms a positive voltage node (3),
first and third semiconductor components (S1, S3) connected in series between the first input terminal (1) and positive voltage node (3), the midpoint between the series connection forming a first node (4),
series connection of first inductive component (L1), first diode (D1) and second inductive component (L2), which series connection is connected between the second input terminal (2) and first node (4), the polarity of the first diode being such that it allows a current to pass from the direction of the second input terminal; the point between the first inductive component (L1) and the first diode (D1) forming a second node (5) and the point between the first diode (D1) and the second inductive component forming a third node (6),
second diode (D2) connected between the second node (5) and the first node (4) in a manner which allows current to pass from the direction of the second node (5) to the first node (4),
third capacitor (C3) connected between first input terminal (1) and the third node (6),
second and fourth semiconductor components (S2, S4) connected in series in parallel with the series connection first and third semiconductor components (S1, S3),
third inductive component (L3), the first end of which is connected to a fourth node (7) formed between the second and fourth semiconductor components (S2, S4) and the second end of which produces first output terminal (8), the second output terminal (9) being formed of the first input terminal (1),
the first and the third semiconductor components (S1, S3) being arranged to control the voltage (Uin) between the first and second input terminals (1, 2).
2. A converter according to claim 1, characterized in that the first and the fourth semiconductor components (S1, S4) are diodes or switch components and second and third semiconductor components (S2, S3) are switch components.
3. A converter according to claim 1 or 2, characterized in that the converter further comprises means for controlling the semiconductor components
(S1, S2, S3, S4).
4. A converter according to claim 1, 2 or 3, characterized in that the means for controlling the first and third semiconductor components is arranged
to control the voltage between the first and second input terminals based on the input
current.
5. A converter according to claim 4, characterized in that the converter comprises a maximum power point tracking device adapted to provide
a voltage reference for the input voltage.
6. A converter according to any one of the previous claims 1 to 5, characterized in that the semiconductor component pairs (S1, S3 and S2, S4) receive the same control.
7. A converter according to any one of the previous claims 1 to 5, characterized in that the semiconductor component pairs (S1, S3 and S2, S4) receive different control and the semiconductor components S2 and S4 are adapted to control the voltage of the series connection of first and second capacitors
(C1, C2).
8. A converter according to claim 7, characterized in that the converter further comprises a controlled inverter (S5, S6, S7, S8) connected to the output terminals of the converter for feeding power to a single
phase alternating grid,
means for determining the polarity of the voltage of the grid (105), means for producing a signal representing the pulse shape of the rectified grid voltage (105),
means for determining the peak voltage of the grid voltage (105), means for controlling the voltage of the sum of the first and second capacitor such that current can be fed to the grid,
means for controlling semiconductor components (S2, S4) such that the output current from the controlled inverter is in phase with the grid voltage.
means for determining the polarity of the voltage of the grid (105), means for producing a signal representing the pulse shape of the rectified grid voltage (105),
means for determining the peak voltage of the grid voltage (105), means for controlling the voltage of the sum of the first and second capacitor such that current can be fed to the grid,
means for controlling semiconductor components (S2, S4) such that the output current from the controlled inverter is in phase with the grid voltage.
9. A method for controlling the converter according to any one of the previous claims
1 to 8, characterized in that the method comprises
measuring the input current (iin) and input voltage (uin),
generating a voltage reference (uin-ref) for the input voltage, and controlling semiconductor components (S1, S3) in response to the measured voltage and voltage reference.
measuring the input current (iin) and input voltage (uin),
generating a voltage reference (uin-ref) for the input voltage, and controlling semiconductor components (S1, S3) in response to the measured voltage and voltage reference.
10. A method for controlling the converter according to claim 8, characterized in that the method comprises
measuring the input current (iin) and input voltage (Uin),
generating a voltage reference (uin-ref) for the input voltage, controlling semiconductor components (S1, S3) in response to the measured voltage and voltage reference,
measuring the grid voltage (Ugrid),
rectifying and scaling the grid voltage,
measuring the voltage (Uim) of the series connection of first and second capacitor,
producing a reference value (Uim-ref) for the voltage of the mentioned series connection of capacitors,
measuring the current (i|AC|) of the output of the converter, controlling the semiconductor components (S2, S4) based on the measured voltage (uim) of the series connection of the capacitors, the reference value (uim-ref) for the voltage of the mentioned series connection of capacitors, the rectified and scaled grid voltage (k|ugrid|) and the measured output current (i|AC|) of the converter for producing output current from the controlled inverter that has the same phase and shape as the grid voltage.
measuring the input current (iin) and input voltage (Uin),
generating a voltage reference (uin-ref) for the input voltage, controlling semiconductor components (S1, S3) in response to the measured voltage and voltage reference,
measuring the grid voltage (Ugrid),
rectifying and scaling the grid voltage,
measuring the voltage (Uim) of the series connection of first and second capacitor,
producing a reference value (Uim-ref) for the voltage of the mentioned series connection of capacitors,
measuring the current (i|AC|) of the output of the converter, controlling the semiconductor components (S2, S4) based on the measured voltage (uim) of the series connection of the capacitors, the reference value (uim-ref) for the voltage of the mentioned series connection of capacitors, the rectified and scaled grid voltage (k|ugrid|) and the measured output current (i|AC|) of the converter for producing output current from the controlled inverter that has the same phase and shape as the grid voltage.